Xing Hang,Dou Zengpei,Xiao Zibing,Xiao Jinxin

Beijing FLUOBON Surfactant Institute,China

Abstract

The environmental and safety issues of perfluorooctane sulfonyl fluoride/perfluorooctane sulfonic acid/perfluorooctane sulfonates (PFOS) were introduced in detail by summarizing several published reviews,including their production,occurrence and emission in environment,exposure and intake to human being,supervision and emission reduction,long-distance transport,bioaccumulation and excretion,degradation and toxicity.Opinion with respect to safety issues of PFOS was also proposed in view of surfactant working field in order to upgrade their application value.The overreaction in the current state for the fear of “fluorine” was concerned.For fire-fighting foams,the issue of “safety” needs to be critically weighed.

Key words

fluorocarbon surfactant;fluoropolymer;PFOS;POPs;PFOA

Fluorinated surfactants and fluoropolymers both contain fluorocarbon chains.They have been applied in many special industrial,daily and high-tech fields due to their high performance.

One of the most important advantages of the fluorocarbon chain is its high thermal and chemical stability.However,such high stability also causes problems in the environment.Substances containing long fluorocarbon chains possess the characteristics of persistent organic pollutants (POPs),namely,environmental persistence: The highly stable fluorocarbon chain is difficult to chemically degrade or bio-degrade through enzymatic and metabolic processes;they are capable of transportation or longdistance migration;they are bioaccumulative in the food chain and have a long half-life in human blood;they are toxic to animals and humans.Therefore,people have paid more and more attention to the environmental and safety issues of fluorinated compounds.

In 2009,perfluorooctane sulfonyl fluoride,perfluorooctane sulfonic acid and their salts (collectively known as PFOS) were listed as POPs by the Stockholm Convention on Persistent Organic Pollutants.[1,2]As for the PFOS issues,environmentalists,toxicologists,surfactant workers,manufacturers and industrial application departments have raised a global heat of research and discussion,[1-11]and even argument,[12]from their own point of view.The environmental and toxicological literature on PFOS has expanded in a speed of over 400 papers per year[1,8]since the first report revealing its being discovered in global range.As a result,currently the opinions of environmental scientists and toxicologists dominate,which cause a global panic about the production and application of PFOS,so that the public has some misunderstanding.We have previously introduced the production routes and the properties of fluorinated surfactants and fluoropolymers and their applications in industry,technology and consumer products.[13]

It should be pointed out that the amount of relevant literature about PFOS is huge,so the present paper will only introduce some important,relatively consistent conclusions in the literature,while some research results that are too specific or currently controversial will not be discussed.

The problems of PFOS

PFOS is an abbreviation for fluorinated compounds involving perfluorooctane sulfonyl fluoride (C8F17SO2F),perfluorooctane sulfonic acid (C8F17SO3H) and their salts or derivatives.In recent years,the substances that can generate PFOS through biological degradation have also been classified as PFOS.Some literature and institutions also call such fluorinated compounds as PFOS-related substances or precursors of PFOS.Later,some environmentalists have expanded the list of PFOS to include compounds containing eight perfluorinated carbons (C8F17-),in which these substances are abbreviated as “perfluorooctylated substances (PFOS)”,which somewhat causes confusion.

In addition,PFOS is the representative for the compounds containing eight perfluorinated carbons (namely C8).The telomers (C8F17CH2CH2I,C8F17CH2CH2OH,etc.) and their derivatives are also the members of C8 family.The members of C8 compounds also involve perfluorooctanoic acid and its salts (PFOA) and any substances that can degrade to PFOA.

Both fluorinated surfactants and fluoropolymers are poly- or perfluoroalkylated substances (abbreviated as PFAS).

Because their bioaccumulation and likely also health risks decrease with decreasing fluorocarbonchain length,PFAS were formally divided into long- and short-fluorocarbon-chain compounds by environmentalists and toxicologists.The long PFAS refers to fluorinated alkylcarboxylic acids (PFCA) with 7 and more fluorinated carbons (e.g.PFOA,perfluorononanoic acid (PFNA)) and fluorinated alkane sulfonates (PFSA) with 6 and more fluorinated carbons (e.g.perfluorohexanesulfonic acid (PFHxS),PFOS),and substances that have the potential to degrade to long-chain fluoroalkyl acids (PFAA) or salts.

Because a large number of experiments have shown that PFHxS and PFOA also possess the characteristics of POPs[2,3](although being weaker than PFOS),such classification mentioned above is more reasonable.PFHxS may be expected to be included in the list of “Stockholm Convention” in the near future.

The environmental and safety issues of PFOS can date back to the 1960s.The presence of PFOS in human blood serum had been surmised in the 1960s,[14]but at that time,analytical methods were not yet capable of identifying these compounds.The interest in the fate of fluorinated organic chemicals in the environment emerged in the 1980s-1990s,when chlorofluorocarbons were banned (i.e.the Montreal Protocol on Substances that Deplete the Ozone Layer) due to their extreme potential for destroying the ozone layer.[8]PFOS is found to be environmentally ubiquitous,such as in the surface water,sediments,wildlife,[15]human body,[16]etc.(PFOA and PFOS have been detected in human blood at μg/L levels from populations throughout the world).[3,6]More concerns connected with fluorinated chemicals have arisen within the last decade.

In 2001,Hansen et al.[5,17]report the presence of several PFAS (mainly PFOS,PFOA and PFHxS) in the sera of nonindustrially exposed humans based on the quantitative results from a study of 65 human sera samples.For the 65 samples reported therein,PFOS was present at the highest concentration.Now it has been verified that the long-fluorocarbonchain PFAS are persistent in the environment,leading to ubiquitous exposure of biota and are bioaccumulative.

The production of PFOS and its emissions and presence in the environment

It is difficult to accurately trace the history of the production of PFAS because most of them are commercially confidential.Large-scale production of PFAS started since the 1970s.3M was,prior to the phase-out announced in 2000,the major global producer of PFOS.[7]3M began to produce perfluorinated compounds by electrochemical fluorination (ECF) in 1949.Excluding the 3M plants in Belgium (Antwerp) and North America (Decatur,Alabama),6 plants were located in Europe (4 in EU member states),6 in Asia (of which 4 were in Japan) and one in South America.From 1966 to the 1990s,production and use grew due to their unique properties.The applications ranged widely including: inks,varnishes,waxes,aqueous film-forming foams (AFFF),metal plating and cleaning,coating formulations,lubricants,water and oil repellents for leather,paper and textiles.Ongoing uses within the European Union in 2004 included: metal plating,photographic,photolithography,and semiconductor industries,hydraulic fluids,and while current stocks last,in PFOS-based AFFF.

Paul et al.have reported that,[7]from 1949 to 2002,the total historical worldwide production of perfluorooctane sulfonyl fluoride (POSF) was estimated to be 96,000 t (or 122,500 t,including unusable wastes).Merely in 2000,3M manufactured approximately 3,665 t POSF,or ～78% of the estimated global POSF production.[7]After 3M's phasing-out its production of POSF,the remaining producers have met the demand of the market and the production is likely to be～1,000 t a-1from 2002 onward.[7,18]

The eight leading chemical companies (Arkema,AsahiGlass,Ciba,Clariant,Daikin,DuPont,Solvay-Solexis,3M/Dyneon) have their own products of PFAS.Indeed,various commercially available compounds have been marketed under the Surlyn®,Capstone®,Fluowet®,Unidyne®,Zonyl®,polyFox®,and Solvera®trademarks.[18]

PFAS have been released to the environment from a series of manufacturing and use point sources (i.e.,to air and water) and more diffusely via widespread consumer use and disposal (i.e.,landfill and incineration).Occurrence of PFOS in the environment arises from two sources: “direct” source,defined as releases during manufacture,application and use of PFAS,and “indirect” source,defined as releases of PFOS as chemical impurities formed during manufacture of POSF,or by breakdown in the environment from POSF-derivatives to PFOS.[7]

Aiming at the data of emissions of PFAS,one should first know the exact data of production of PFAS,which,however,is very difficult,or even impossible,to obtain.Estimation could be made based on some public information if certain assumption were introduced.Based on the public information of 3M.,the production information from 3M's Director of Regulatory Affairs,cited in Olsen et al.'s work,[19]the report published by the Organisation for Economic Co-operation and Development (OECD) and the estimation by Smithwick et al.,[20]Paul et al.have estimated the global historical production for POSF,the global historical environmental releases of PFOS and their trends with time.[7]According to Paul et al.'s estimation,the total historical worldwide production of POSF was estimated to be 96,000 t (or 122,500 t,including unusable wastes) between 1970-2002,with an estimated global release of 45,250 t to air and water between 1970-2012 from direct and indirect sources.Estimates indicate that direct emissions from POSF-derived products are the major source to the environment resulting in releases of 450-2,700 t PFOS into wastewater streams,primarily through losses from stain repellent treated carpets,waterproof apparel,and aqueous fire fighting foams.[7]The properties of PFOS (high water solubility,negligible vapor pressure,and limited sorption to particles) imply it will reside in surface waters,predominantly in oceans.Measured oceanic data suggests～235-1,770 t of PFOS currently reside in ocean surface waters.[7]Environmental monitoring from the 1970s onward shows strong upward trends in biota.Since cessation of POSF production by 3M in 2002,a reduction in some compartments has been observed.

A comprehensive quantitative inventory of global production and emissions of C4-C14 PFCA over the 1951-2030 period has been estimated byWang et al..[9]In a companion paper[10]they discuss sources that are known to exist but cannot be quantified presently for lack of information.Krafft et al.have summarized the data mentioned above and further provide a review on the production and emissions of long-fluorocarbonchain PFSA.[2,3]According to their estimation,the total emissions (1951-2015) of PFCA (mostly PFOA and PFNA-based products) were estimated to be 2,610 to 21,400 tons.[3,9]They predict that projected emissions between 2015 and 2030 (mainly from PFOA use in China and Russia,and from degradation of long-fluorocarbon-chain FTOH (fluorinated telomer alcohol)-based products) are from 20 to 6,420 tons.[3,9]

Moreover,the ubiquitous presence of longfluorocarbon-chain PFAS continues to be documented,including in food and drinking water,dust,wildlife and humans.[3]

Environmental PFAS distribution is not uniform.While open ocean water PFOA levels are typically in the 10～80 pg/L range (1～20 pg/L for PFOS),these levels can be several orders of magnitude higher in industrialized areas,coastal waters,or downstream from an airport.[3]Drinking water PFAS levels are usually low and considered safe.Fresh water PFAA concentrations can,however,be higher near a direct industrial source or near a fire-fighting training area.[3]

It is noteworthy that AFFF is one of the most important emission sources of PFOS.In many countries,elevated PFAS levels are found in the vicinity of airports and fire-fighting training centers because of the accidental spills and training exercises.For example,training exercises occurred on a weekly to monthly basis at Naval Air Station (NAS) Fallon,Nevada,and used 1200-3200 L (3%～6%) of aqueous AFFF solutions per week.[21]Many fluorinated surfactants (C4～C12) are present in AFFF.No less than 103 PFAS (fluorocarbon-chains from C3to C15) were detected in 10 firefighting products.[3,22]Many of them were measured in above average amounts in the blood of fire fighters,including PFOS and PFHxS.[3]

Carpets are also an important emission source especially in European countries,in which the fluorinecontaining repellents will slowly release over time.Fluorinated telomer alcohol (abbreviated as FTOH) (primarily 8:2 FTOH) and sulfonamidoethanols (abbreviated as FOSE) (primarily C8F17SO2N(CH3)CH2CH2OH (abbreviated as N-MeFOSE)) and their derivatives are found in high levels near carpet production and treatment facilities,and in consumer goods.

Discharges from wastewater treatment plants (WWTP),fluorochemical production sites and landfills,amendment of soil with industrially impacted biosolids from WWTP and urban runoff waters also participate to dissemination of and exposure to PFAS.[3]

PFAS adsorb strongly onto suspended solid particulates and sediments,especially the longerfluorocarbon-chain (C12-C15) PFCA.PFOA was shown to interact with soil-bound proteins.[3]PFAS adsorb to soil,sediments and sewage sludge,and subsequent release to surface and ground waters and into crops.It is unclear why PFCA and N-MeFOSE,and PFSA adsorb differentially on smaller and larger particles,respectively.PFOS may form complexes with clay through hydrogen bonds at the mineral's surface,while PFHxS would bind with the inner OH.[3]

Many long-fluorocarbon-chain PFAS come from the atmospheric and biological degradation of fluorinated surfactants and fluoropolymers (they can yield mainly long-fluorocarbon-chain PFAA after degradation),which may be an important source of indirect emission of PFCA.

Exposures and intakes of PFOS

The presence of PFAS in biota is ubiquitous and no one can escape exposure.Figure 1 shows the data of mean serum concentrations (μg/L) of PFOS,PFOA and PFNA for the general population reported by US National Health and Nutrition Examination Survey (NHANES).[3]As shown in Figure 1,the supervision of US on PFOS and PFOA (including 3M's phasingout PFOS) is quite effective.

Figure 1.Geometric mean mass concentrations of PFOS,PFOA and PFNA in human serum[3]

Exposures are not uniform: some are more exposed than others,or more vulnerable.The exposure increases with trophic levels,maximizing for the predators sitting at the top of the food chains,to which humans belong.However,exposure and the associated risk are clearly not the same for the general (background) population and for those more exposed,which includes workers producing or using PFAS and populations living near a source of contamination.Serum levels can be several orders of magnitude higher in occupationally exposed than in background populations

Biomonitoring of identified higher exposed populations has recently been reviewed by Olsen et al..[23]Fluorochemical plant workers,ski waxing technicians,employees of fisheries located near fluorochemical producer and user plants,fire fighters (exposed to AFFF) can show PFOS and PFOA serum concentrations much larger (several times to 2-3 orders of magnitude) than the general population.The highest ever reported were 114,000 μg/L for PFOA and 12,800 μg/L for PFOS.[3,23]More exposed resident populations are those living near a point source of PFAS,as through polluted drinking water.

For the general population the major source of contamination arises from food and sometimes drinking water (＞90%).PFOS is generally the most abundant PFAS found in humans,the second usually being PFOA,and PFHxS the third.Exposure to PFOS and the higher PFCA (n ≥ 9) through consumption of fish,in particular from polluted waters,can be very large.PFAS migrate easily and in substantial amounts from food packaging into food.PFAA levels in drinking water are usually low,unless near a point source of PFAS.PFOA serum levels can then increase significantly with on-going drinking water exposure.Cows,and hence,dairy milk are not sheltered against PFAS pollution.[3]An advanced estimation (2012) of dietary exposure concluded that human dietary exposure to PFOS and PFOA is highly unlikely to exceed health-based guidance values.[3]

Human exposure also arises from indoor and ambient air and house dust.Outdoor air carries volatile FTOH,C8F17SO2NH2(abbreviated as FOSA),FOSE,etc.Non-volatile PFAS,particularly PFOA,are adsorbed on airborne particles.House dust (mainly from impregnation sprays and impregnated carpets) can be a significant contributor to exposure to the less volatile PFAA.Up to 27 different PFAS were detected in a study of house dust.FTOH precursors,including polyfluoroalkyl phosphate esters (PAP),appear to be significant sources of human PFCA contamination.[3]Lesser exposure sources include contact with impregnated clothes,footwear,upholstery,and other goods that have been treated with PFAS or contain unintended residues used during manufacturing.

Substantial sex differences have been observed.Women have age-dependent lower PFOS,PFOA and PFHxS levels than men,due in part to elimination by menstruation.[3]Children,especially toddlers,are more exposed than adults on a body weight basis,due to higher relative food consumption and to handto-mouth transfer from impregnated carpets and ingestion of dust,resulting in higher PFAS serum levels than for adults.

In utero and postnatal exposures are of particular concern.PFAA cross the placental barrier,exposing neonates via their mother's blood.[3]Numerous PFAS have been detected in breast milk,though in concentrations much lower than in the mother's blood plasma.Lactational transfer (and baby formulations) can nevertheless be the primary source of exposure to PFAS,mainly PFOS,for infants.[3]

Supervision and emission reduction of PFOS

Historically,“land farming” and landfilling were used to dispose of wastewater sludge and solids,while wastes and off-spec products were disposed to water.Both land farming and off-spec aquatic discharges were apparently discontinued by 1998.[7]Around the same time,a discotherm fluorochemical capture and reuse system was added to the manufacturing plants,reducing releases to wastewater,while an activated carbon adsorption system was added to the wastewater treatment works at both Decatur,Alabama and Antwerp,Belgium.Carbon adsorption has been shown to be effective at removing both PFOS and POSFderived compounds,while designated wastewater streams were collected and transported off site for disposal via hazardous waste incineration.[7]Together,these systems reduced emissions to air and water by 40 and 50%,respectively.On the basis of Decatur's 90% solid waste stream,63% of fluorochemical waste is estimated to have been disposed of through incineration,33% was sent to hazardous waste landfill and 4% to nonhazardous waste landfills.[7]The only effective way of totally destroying POSF is through high temperature,hazardous wastes incineration.[7]

The global distribution of PFOS and their precursors in biota and humans has made them become the very concerned chemicals.Further use of long-chain PFAS that have environmental persistence,bioaccumulation and possible health risks is environmentally not sustainable.Under the pressure from environmentalists and toxicologists,as well as the public concern about the ubiquity and toxicity,international communities have decided and implemented corresponding regulations and voluntary actions.

PFOS began to be gradually phased out in the early 2000s[5]and its emission reduction was started by 3M who is globally the biggest producer of PFOS.In 2000,under the guidance of the US Environmental Protection Agency (USEPA),the 3M Company started phasing out its production of PFOS,PFOA and related compounds.3M also announced at that time that they would voluntarily phase out PFOS and similar chemicals by 2002.Moreover,in 2006,the USEPA teamed up with the fluoropolymer industry to launch the global “Voluntary PFOA Stewardship Program”,with the long-term goal of eliminating PFOA emissions altogether by 2015.[5]The USEPA then established a Stewardship Program involving eight major chemical companies in Western Europe,the United States and Japan that 1) aimed at a 95% reduction by 2010 of facility emissions of PFOA precursor chemicals that can break down to PFOA and related higher homologues,and of content levels of these chemicals in final products;and 2) committed to working toward the elimination of these chemicals from emissions and products by 2015.Another voluntary action led to virtual elimination of PFOS use in semi-conductor production.[3]

It is worth mentioning the event of “DuPont Teflon” in the US in July 2004.Although this event should be ascribed to ammonium perfluorooctanoate which is not a PFOS,people began to highly focus on the pollution and risks possibly caused by fluorinated organic compounds (PFC) (especially the widely applied ones such as PFOS and PFOA).Finally,PFC have been quoted as the “PCBs of the XXIst century”[18]and PFOS was regarded as POPs,and PFOS and its precursor POSF were listed in Annex B of Stockholm Convention in May 2009.

Figure 2 provides a brief history of PFAS production and regulation as summarized by Krafft et al.[3]

Figure 2.Brief history of PFAS production and regulation[3]

Besides the “Stockholm Convention”,different countries and organizations have also implemented regulation and restriction on other PFAS.[3]The Organization for Economic Cooperation and Development (OECD) lists perfluorinated C5～18compounds,which includes most perfluorinated surfactants,as high-production-volume (HPV) chemicals.High-production-volume chemicals are those chemicals manufactured in or imported to the US in volumes exceeding 1 million pounds.[21]

In 2008,the European Food Safety Authority (EFSA) proposed the provisional tolerable daily intakes (pTDI) values of 150 ng/kg body weight/day for PFOS and 1500 ng/kgbody weight/day for PFOA on the basis of animal studies.The TDI for PFOS is generally above the indicative figure estimated by EFSA for human exposure,but could be exceeded for the highest exposed populations (as by high fish consumption,or occupational exposure).For PFOA,the TDI is well above the indicative value for average human exposure,although lower pTDI may be desirable for children).[3]

Establishing guidelines for drinking water is also critical in terms of public health protection.Maximum tolerable values differ somewhat depending on country and agency and range from 0.04 to 0.5 μg/L for PFOA and 0.2～0.3 μg/L for PFOS.In the US,six PFAS have been listed in the proposed third Unregulated Contaminant Monitoring Rule (2012-2016) for drinking water safety,and their minimum reporting levels (μg/L) were set.Concerning the shorter fluorocarbon-chain acids,the provisional healthrelated indication values for drinking water to be considered safe for life-long exposure (μg/L) were much larger than the long ones.Water quality criteria are also being set in China.[3]

In the international activites of phasing-out PFOS,specific exemptions and acceptable purposes were set in view of the irreplaceable application value in some areas and the economic benefit brought by PFOS.The deadline for use in these areas is different for different countries,especially concerning the benefits of developing countries.Specific exemptions include photo masks in the semiconductor and LCD industries,hard metal plating,decorative metal plating,electric and electronic parts for some color printers and color copy machines,insecticides for control of red imported fire ants and termites,chemically driven oil production,etc.Acceptable purposes include photo imaging,photo resist and anti-reflective coatings for semi-conductors,etching agent for compound semi-conductors and ceramic filters,aviation hydraulic fluids,metal plating only in closed-loop systems,certain medical devices (e.g.ETFE layers,radio-opaque ETFE,in vitro diagnostic medical devices,CCD color filters),fire fighting foam,etc.

It should be mentioned that,recently,in the “Fourteenth meeting of the Persistent Organic Pollutants Review Commtitee of the Stockholm Convention”,the committeesuggestthat some acceptable purposes and specific exemptions may be altered according to the current situation of PFOS and replacement products (Reference No.: POPRC-14/3),as listed below:

(1) Theysuggestthat,photo imaging,photo resist and anti-reflective coatings for semi-conductors,etching agent for compound semi-conductors and ceramic filters,aviation hydraulic fluids,and certain medical devices (e.g.ETFE layers,radio-opaque ETFE,in vitro diagnostic medical devices,CCD color filters) are no longer acceptable purposes;

(2) Metal plating only in closed-loop systems,and fire fighting foam are bothsuggestedto be changed from acceptable purposes to specific exemptions (P.S.the specific exemption of PFOS in fire-fighting foams is only allowed for Class B fire);

(3) Insect baits for control of leaf-cutting ants is still acceptable purpose but the clause issuggestedto be revised,i.e.,the Sulfluramid (CAS No.4151-50-2) in insect baits,as an active component for control of leaf-cutting ants,is only acceptable for agricultural use.

In addtion,for the specific exemptions,the committeesuggestthat,photo masks in the semiconductor and LCD industries,hard metal plating,decorative metal plating,electric and electronic parts for some color printers and color copy machines,insecticides for control of red imported fire ants and termites,and chemically driven oil production are no longer specific exemptions.

It is worth noting that the use of PFAS in some areas are confidential and unkown to public,so are currently not under the list of phasing-out/Specific exemptions/Acceptable purposes.With the development of high and new technology,these applications will be known and request specific exemptions.

Although regulations are necessary to control the pollution and risks of PFAS,the regulations have their own limitations and weakness,including the adverse effects on economic development.

Long-distance transport of PFOS

Researchers are confused about the ubiquitous distribution of PFOS.PFOS has even been discovered in arctic areas that are far away from the places of production and use of PFOS.PFOA and PFOS accumulations in marine mammals (e.g.,＞1200 ng/g wet weight in liver) are common all over Northern hemisphere while levels are much lower in the southern hemisphere.[18]How can they be transported to places such as arctic areas? PFCA and PFSA are persistent,and they can be biomagnified in the aquatic or marine food chains.They are moderately water soluble,and their vapor pressure are very low or negligible,which implies that it is impossible for PFAA to distribute into the atmosphere so that PFAA have no remarkable characteristics for long-distance atmospheric transport.POSF is volatile,but most of the fluorinated surfactants (especially ionic ones) produced from POSF are non- volatile.There is a common view in current academia,that is,although PFOS is difficult to be degraded or metabolized,the precursors of PFOS may be degraded or metabolized through hydrolysis,photolysis,microorganism,etc.and these products of degradation or metabolism can be transported long distances.Long-range dissemination involves both an aquatic (oceanic) route that is favored for the more water soluble compounds (e.g.fluorinated alkyl acids and their salts),and an atmospheric route for the more volatile,neutral ones (e.g.fluorinated telomer alcohols (n:2 FTOH),sulfonamidoethanols (FOSE),phosphate diesters (diPAP)).The latter have atmospheric lifetimes of tens of days and are eventually degraded to PFAA that can then be deposited in remote places.[3]The anionic species are preferentially transported in solution by oceanic water currents,in surface foam and aerosols.An estimation suggested pKavalues of 0.5 for PFOA and below 0.3 for PFSA.[3]The relative contributions of the two routes are still being debated.[3]Polar ice melting consecutive to global warming may allow release of atmospherically deposited PFOA to the oceans.[3]

Bioaccumulation and excretion of PFOS

One of the biggest concerns about PFAS is whether they are bioaccumulated in individuals and biomagnified in the food chains,leading to maximizing for the predators sitting at the top of the food chains.

Bioconcentration (direct uptake of chemical from its environment) and bioaccumulation (uptake from all exposures,including food) of PFAS in animals increase steeply with fluorocarbon-chain length.For a given fluorocarbon-chain length,bioaccumulation is more effective for PFSA than for PFCA (e.g.PFOS ＞PFNA),possibly because of tighter binding to proteins.[3]Bioaccumulation depends also on species,with perfluorobutane sulfonate (PFBS) and PFOS mainly found in crabs,PFOS and perfluorodecanoic acid (PFDA) in fish,and PFHxS in gastropods and bivalves.[3]Biomagnification of PFAS (increased pollutant concentration in predator vs.prey) is observed when moving up within food chains and trophic levels,and is particularly effective for PFOS and the longer-fluorocarbon-chain PFCA,in top predators such as bear or wild mink.Very long,up to C15PFCA were detected in fish and eggs of sea birds.[3]Transfer of PFOS from zebra fish to their eggs has been established.[3]The distribution and monitoring of PFAA in wildlife has recently been reviewed.[3]

The shorter,alternative compounds (≤C4 for PFSA- and ≤C6 for PFCA-related compounds) that are now being used have bioconcentration factors (BCF) far below the upper limits set by the EU (BCF＜1000) or USEPA (BCF＜2000) and are not considered bioaccumulative.[2]These compounds are,however,as persistent in the environment as their higher homologues.

Animal (including human) data show widely different pharmacokinetics,including elimination rates.Pharmacokinetics and (eco)toxicological investigation results are characterized by wide product,species,gender and other variations that are far from being fully understood.Oral PFAS absorption is generally rapid.Distribution occurs primarily in the liver,plasma and kidney,but depends somewhat on compound,animal and type of exposure.The liver,lung and kidney are major target organs for PFAA.

No instance of metabolism has yet been reported for PFAA.Metabolism of FTOH-based compounds in animals or by animal tissues involves many of the pathways and metabolites identified for microbial degradation by sludge,soil or microbial cultures.There are important differences between the processes in animals and microbial systems.

The body half-life increases with the fluorocarbonchain length for PFCA,which parallels that observed for fluorocarbons,but these half-lives are about two orders of magnitude higher,for the same fluorocarbon-chain length.Fluorocarbons benefit from an excretion route,exhalation with expired air that is clearly not available for PFAA.

The persistence of long-fluorocarbon-chain PFAA in human body is very long,with half-lives being several years as reported by most of the literature.Owing to the testing samples,specimen,dosage,etc.,the half-lives of PFAS vary in a wide range in different literature reports.For PFOS,a biomonitoring study of Red Cross blood donors yielded a half-life value of 4.3 years.A larger halflife of 8.2 years was derived from the decline in PFOS serum levels observed in Swedish women.Fitting a pharmacodynamics model to US NHANES data sets led to estimated PFOS elimination halflives of 4.7 years for men and 3.7 years for women,menstruation accounting for ～30% of the difference.[3]References [3,18,24] have reported the data of halflives in human serum: PFOA (3.6 years),PFOS (5.4 years),PFHxS (8.5 years).For the fluorocarbon chain as short as C4,Olsen et al.find that,[25]compared to PFOS and PFHxS,PFBS has a much lower potential for accumulation in human serum.PFBS is eliminated at a greater rate from human serum than the higher chain homologs of PFOS and PFHxS.[25]The geometric mean serum elimination half-life of PFBS was 25.8 days (95% confidence interval 16.6-40.2 days) compared to a PFOS geometric mean of 1751 days (95% confidence interval 1461-2099 days) and a PFHxS geometric mean of 2662 days (95% confidence interval 2112-3555 days).[25]The puzzling and unexplained out of line longer half-life of PFHxS in humans,superior to that of its higher homolog PFOS,[3]was confirmed in a study that also noted huge differences between young women and men and older women,with mean half-lives in young females of 6.7 and 7.7 years for PFOS and PFHxS,respectively,versus 34 and 35 years for males and older females.[3]The much slower excretion rate observed for humans as compared to animals was assigned to larger renal saturable resorption efficiency.[3]

The metabolism and pharmacokinetics of PFAS have recently been reviewed by Kudo et al.,[3,26]in which the plasma or serum half-lives of PFAS with different fluorocarbon-chain length in animals and humans are reported in detail.Krafft et al.have also compared these literature data in their review.[3]

Degradation of PFOS

There are many systematical reviews about the degradation of PFOS.[3,8,27]Frömel et al.have systematically reviewed the biodegradation of FTOH and fluorinated telomer olefins,as well as their derivatives and those fluoropolymers produced from them as inermediates,and N-alkyl perfluorooctane sulfonamide and its derivatives.[8]Young et al.have detailedly discussed the atmospheric chemistry,degradation pathway and lifetime of PFAA precursors and the PFCA produced from these precursors,as well as their sources and content levels.[27]

PFAA (e.g.PFOA,PFNA,PFOS,PFBS) are extremely refractory to degradation and there is still little definite evidence for their degradation in environmentally relevant conditions.Direct photolysis of PFOA is,for example,barred since it does not absorb light from the solar spectrum.Indirect degradation by hydroxyl radicals produced in the presence of natural sensitizers (e.g.nitrate and ferric ions and seawater) was not detected when PFOA was irradiated in the 290-800 nm range.PFOA was,however,degraded (98% in 28 days) in the presence of Fe(III) under sunlight.[3]Radical reactions involving UV-induced electron transfer from PFOA to Fe(III) or attack by hydroxyl radicals were proposed.[3]Taniyasu et al.'s paper suggesting that photolysis of long-fluorocarbon-chain PFSA and PFCA was possible in environmental conditions,and that the shorter fluorocarbon-chain homologues were more resistant to photodegradation,has been severely criticized by Wang et al..[3]Interestingly,PFOA has been successfully degraded to shorter PFCA and F-using a horseradish peroxidase (an iron(III) hemecentered enzyme)-catalyzed reaction with hydrogen peroxide and a co-substrate,4-methoxyphenol,a reaction that may have potential for the treatment of PFCA-contaminated waters.[3]

There are a lot of studies on the atmospheric degradation of PFAS.Atmospheric degradation ofn:2 FTOH-derived surfactants and polymers in atmosphere-mimicking conditions has been intensely studied.[27]An “unzipping” cycle that involves fluorinated alkoxy,fluorinated alkyl and fluorinated alkyl peroxy radicals,and splitting off of COF2yields shorter fluorocarbon-chain PFCA homologues.[3]

No definite evidence of biodegradation of PFAA has yet been produced.Biodegradation of long-chain fluorinated telomer alcohol and fluorinated alkane sulfonamide derivatives and precursors by various bacteria and by microorganisms present in activated sludge,soil and sediments,as well as by hepatocytes and microsomes in vitro,and in animals,has been confirmed.[3]The routes of biodegradation for some important PFAA and PFOS precursors are shown below.

Long-fluorocarbon-chain FTOH

FTOH play a pivotal role in the degradation processes of many commercial products (e.g.esters,PAPs,acrylate,metacrylate monomers and polymers),which first hydrolyze to FTOH.Outcome,rates and yields of aerobic microbial degradation ofn:2 FTOH depend critically on fluorocarbonchain length,isomer,bacterial strain,environmental matrix (type and origin of soil,sediment,WWTP sludge,etc.),co-metabolites and testing conditions,among others,differences in microbial and fungal populations and enzyme efficacy and specificity.n:2 FTOH (n=4,6,8) were biodegraded byPseudomonasbacterial strains,leading to removal of CF2units and production of short-fluorocarbon-chain acids.Aerobic biodegradation in soil of 8:2 FTOH stearate ester and citrate triester surfactants has been established.Frömel et al.have described the pathways of degradtion of 8:2 FTOH in their review.[8]The biodegradation of 8:2-FTOH begins with the oxidation of the hydroxyl function to the respective aldehyde (8:2-FTAl) (F(CF2)8CH2CHO) and to the carboxylic acid (8:2-FTA) (F(CF2)8CH2COOH).The 8:2-FTA is then transformed to the unsaturated carboxylic acid (8:2-FTUA) (F(CF2)7CF=CHCOOH) by formal cleavage of hydrogen fluoride,representing the next step within the circle of β-oxidation.The pathways mentioned above also explain the phenomenon that is confusing for a long time,that is the odd numbered,and not the even numbered PFCA,were the predominant pollutants.[11]

C8F17SO2N(C2H5)CH2CH2OH (N-EtFOSE)

N-EtFOSE undergoes degradation pathways similar to 8:2 FTOH.[8]N-EtFOSE is rapidly transformed,primarily by the aerobic microorganisms present in activated sludge.Similar to FTOH,the hydroxyl function is initially oxidized to the respective aldehyde and to the carboxylic acid (N-EtFOSAA) (C8F17SO2N(C2H5)CH2COOH),followed by cleavage of the acetic acid moiety,which leads to C8F17SO2NHC2H5(N-EtFOSA),and then successively changes to C8F17SO2NH2(FOSA),C8F17SO2H (FOSI) and finally the persistent C8F17SO3-(PFOS).[8]

Fluorinated telomer ethoxylates

Fluorinated telomer ethoxylates (CnF2n+1(CH2CH2O)pH,n=6～10，p= ～13) were aerobically degraded in WWTP effluent.Fluorinated telomer sulfonate (FTSA) 6:2 FTSA (K+salt) was slowly biotransformed by activated sludge.The “one-carbon removal pathway” for sequential removal of CF2groups in sludge was identified.[3]

Polyfluoroalkyl phosphate esters (PAP)

6:2 mono- and diPAPs,and mixtures ofn:2 monoPAPs (n=4,6,8,10) were microbially hydrolyzed to FTOH,eventually releasing PFCA in WWTP effluents.[3]

Fluorinated monomers

Microbial degradation of C8F17C2H4OC(O)CH=CH2(8:2 FTAc) and C8F17C2H4OC(O)C(CH3)=CH2(8:2 FTMac) monomers in aerobic soil yielded first 8:2 FTOH and then C7F15CH(CH3)OH (7:2 sFTOH),FTCA and PFCA,including PFOA.[3]

Fluoropolymers

Surprisingly little is published about the environmental biodegradation and fate of the commercially important,large tonnage fluoropolymers.Conflicting studies conclude differently.Higher levels of PFOA,PFNA,PFDA and PFOS have often been found in WWTP outflows relative to inflows,also indicating biotransformation of precursors,as well as ineffective removal of PFAS by conventional wastewater treatment.Microbial intervention is also likely involved in the quasi-irreversible binding of PFAS to soil.[3]Similar to fluorotelomer-based polymers,N-EtFOSE and N-MeFOSE-based polymers may be enzymatically cleaved or hydrolyzed to N-EtFOSE and N-MeFOSE,respectively.[8]

It should be emphasized that,“PFAS is difficult to be degraded” is true merely for the perfluoroalkyl groups,while the other parts of the PFSA molecule are still degradable (The degradation products are PFSA or PFCA and their salts).Moreover,it should also be pointed out that,“the fluorocarbon chain is difficult to be biodegraded” merely means that it may take a long time for biodegradation.There are examples indeed for microbial defluorination.[18]For example,F(CF2)7CF=CHCOOH (8:2 FTUA) can be further metabolized by several pathways.[8]One such pathway gives rise to the F(CF2)7CH=CHCOOH (7:3 FTUA) and then to the potentially persistent F(CF2)7CH2CH2COOH (7:3 FTA),which was first described in the literature by Wang et al..If this step can be confirmed,this would be one of the first reports of microbial defluorination.[8]

The biodegradation of chemicals depends on the corresponding microorganisms (bacteria).PFOS are synthetic chemical substances,so no corresponding fluorophage bacteria have been found in nature.Here are two questions: strictly speaking,no fluorophage bacteria has been found in nature does not definitely mean no fluorophage bacteria in nature;According to the law of natural evolution,since PFOS has been detected everywhere in nature now,it can not be ruled out that with the ubiquitous presence of PFOS in nature,fluorophage bacteria may exist (These two points are our hypotheses without any scientific proof).Krafft et al.[3]have imaginations that: “Ideally,one should find or generate and breed microorganisms (“bugs”) that would feed on F-chains and release chocolate.” It may become true in future,but for now：“Alas,this covetable approach has,to our knowledge,not yet materialized.[3]”

Another reason for the fluorinated surfactants being difficult to be biodegraded is the poor hydrophilicity of fluorocarbon chains.It should be noted that,it is the fluorocarbon chains that are highly hydrophobic,not the surfactants themselves.In general,the biodegradation of certain chemicals can be closely related to their solubility in water to ensure sufficient contact with microorganisms (microorganisms usually live in water environment),i.e.,the less soluble,the more difficult to biodegrade.For example,the biodegradation of several kinds of hydrocarbons is obviously accelerated by β-cyclodextrin due to the increase of solubility by formation of inclusion complexes.[28]Whether the degradation of PFAS could be accelerated if the solubility of PFAS were improved for effective contact with corresponding microorganisms? That would be a meaningful question.

There is another misconception that should be avoided.Persistence is one of the properties of POPs,however,it does not mean that any substance being persistent will be a POP.C4-fluorinated surfactants are just the case.Although C4F9-groups are also difficult to biodegrade,C4-fluorinated surfactants are not POPs because they do not accumulate in human body (the half-life is short) and are less toxic.

It is worth mentioning that,as for the degradation of perfluoroalkyl ethers,to what extent can these PFAS be degraded? This will be discussed in detail in the next serial article.

Toxicity of PFOS

It is a controversial issue at present.Up to now,the research and reports on PFOS mainly focus on “finding” PFOS,that is,in which places,which animals,which human organs (parts) PFOS have been found.Currently,the research on the toxicity of PFOS is basically animal experiments,such as toxicological research of PFOS and PFOA on monkeys,rats,mice,rabbits,etc.This is also a general procedure for studying the toxicity of chemicals,which cannot be tested directly on humans.

In US,an epidemiological study (～69,000 participants enrolled),i.e.the C8 Health Project (2005-2013),was set up.The project investigated possible links between the elevated PFOA serum levels and a large range of human diseases.[3]The “probable link” reports submitted by the C8 Science Panel to the Court (2011-2012),about whether there was,or not,a probable link between exposure and disease.[3]Krafft et al.have reviewed the literature results in detail,and discussed the exposure and toxicity of PFAS in aspects of different populations (occupationally exposed populations,more exposed populations,and general population),human epidemiological studies,(eco)toxicological investigation and pharmacokinetics (including elimination rates).[3]

Strictly speaking,the current evidence for an association of PFOS exposure with health effects was inconclusive.Overt toxicity or mortality attributable to PFAS has so far not been reported in humans,even at the highest occupational levels.Epidemiological findings have recently been critically analyzed.[3]

There are some results related to the effects of PFOS on human body,but these results are based on statistics,that is,by testing the concentration of PFOS in human blood,and then record the PFOS concentration and some parameters of the human body,such as height,weight,adiposis,mortality of certain disease and so on.The statistics made by 3M Co.on the disease of workers in its electrochemical fluorination plant have shown links of PFOS to bladder and prostate cancers.PFOS can cause the toxication of liver,nerve and immune system,seriously can even cause cancer.In particular,the endocrine system of newborns is more vulnerable,so PFOS will have adverse effects on the health of humans and future generations.PFOS usually does not accumulate in adipose tissue after intake by living things.Most will bind to plasma proteins in the blood,while the rest will accumulate in the liver tissue and muscular tissue of animals.

In 2014,an International Agency for Research on Cancer working group classified PFOA as possibly carcinogenic in humans.Some studies have linked PFOA to infertility,immune problems and developmental defect of fetus.Even in the kitchen,teflon can also cause problems: After heating for a few minutes,when the temperature is over 260 ℃,the teflon coating will decompose and emit toxic fumes.PFOA has contaminated the drinking water near a Dupont plant in West Virginia and PFOA can pass from mother to fetus across the placenta.

Here it should be noted that:

(1) Although there are some reports on the possible links between PFOS and human body parameters,so far,these reports are inconclusive,and some of them are contradictory.

(2) The vast majority of current studies are conducted on animals.However,in many cases,the toxicology is greatly different between animals and humans.Even in animal studies,the results are not entirely consistent.[3]These data show considerable,often inconsistent variability,raising questions about the correlation of animal models to human risk assessment data.Currently,the effects of different PFAS,gender and species,as well as the complicated pharmacodynamic behaviors and toxic mechanisms,are not fully understood or predictable.Some animal data are definitely not applicable to humans.

(3) The distribution of PFAS in animal tissues varies in different species.Currently,the data collected are mainly about PFOS and PFOA,while the pharmacokinetic effects of PFAS on health are widely dependent on compounds,species,gender,etc.The results can be different,which are often inconsistent,sometimes contrary,departured from linearity and usually pointing to different mechanisms.

(4) Statistical results like those from 3M Co.are based on long-term human exposure (workers in chemical plants of electrochemical fluorination),while the general population is unlikely to expose such high concentrations of PFOS.Reactions,no matter chemical or biological ones,are dependent on the concentration of reactants,i.e.reactions can occur only when the concentration of reactant reached a certain level.And there is no conclusion about what concentration of PFOS needs to reach to harm the human body.

(5) Many chemicals may not be toxic,but their degradation products are.For example,trifluoroacetic acid (CF3COOH),when it is turned into HCF2COOH it would be one of the most toxic substances.

(6) Little is known about the possible combined effects of PFAS when mixed with other chemicals,drugs and pollutants.The complicated interactions between different chemicals result in complicated effects on human health.

Although there is no clear conclusion about the risks of PFOS on human body at present,it is correct to restrict and phase out PFOS.Any synthetic chemicals,if they are not confirmed to be harmless to the human body,must be used with caution.

It is worth noting that some positive effects of PFOS have been found in some studies.For example,PFAS exposure has a neuroprotective effect on the elderly people and can reduce the risk of cognitive limitations,especially for diabetic patients.[3]

The voice of “fluorine free” may be misleading

The remediation of PFOS or PFOA in the environment will cost high.Many conventional treatment technologies are ineffective.Kucharzyk et al.[29]have reviewed the treatment technologies for remediation of PFOS- or PFOA-impacted groundwater,such as extraction with carbon adsorption,sonochemistry,bioremediation and photolysis.In some cases,available information on the impact of treatment technologies on precursors of PFOA and PFOS and/or smaller-chained PFAS is also noted.

On the other hand,the fate and transformation of PFAS in landfills are also concerned,indicating that in landfill leachate the short-chain PFAA (C4-C7) are abundant.[30]As mentioned above,the short-fluorocarbonchain PFAS (e.g.PFSA≤C4and PFCA ≤C6) are also persistent,so some people have made effort in developing “fluorine-free” products,despiting the fact that short-fluorocarbon-chain PFAS are nonbioaccumulative and less toxic.

On March 27,2018,the governor of Washington State in the US signed a bill banning the use of PFAS in fire-fighting foams.Noting that this bill will not be applied to airports (fuel storage and delivery facilities) or chemical plants in the US.Even the latest information released by “Stockholm Convention” suggested a signal that phasing-out all PFAS (including the short fluorocarbon-chain ones) in fire-fighting foams and some other applications may be the future trend.However,we DO NOT think it is a good idea.As Mike Willson said: “Many are perhaps encouraged to use fluorine-free foams when transitioning away from undesirable legacy longchain (PFOS/PFOA) fluorinated foam agents...but do they provide the necessary balance between the pollution derived from fire against pollution derived from firefighting?” The fire is threatening human life and property,anytime,anywhere.Owing to the different scientific mechanisms,the extinguishing performance of fluorine-free foams can not be as efficient as PFAS-based ones.Owing to the benifits from fluorination (e.g.oil repellency,high surface activity),the short fluorocarbon-chain PFAS,though still have shortcomings at present,will be more reliable than fluorine-free substances.It's too early to give up the strategy of short fluorocarbon-chain PFAS for replacement of long-fluorocarbon-chain PFAS.Let alone the “risk” of the former is potential,possible and not verified.The fear of “fluorine” is undoubtedly an overreaction.

Summary

PFOS has been regulated and restricted by the international community owing to its character of POPs.On the other hand,its unique characteristics have given it specific exemptions in many areas of application (such as aviation hydraulic fluids,metal plating,medical devices,etc.).Up to now,the research on the environmental and toxicological issues of PFOS is still very inadequate,the conclusions obtained are still lack of sufficient scientific evidence,and the reported data vary in a wide range.Nevertheless,safety is important.The regulation and restriction on PFOS are necessary in the current circumstances.With the development of methodology,we hope that: Various isomers of PFOS compounds can be synthesized and separated;the collection of physical and chemical data is continued;deep-going environmental and toxicological research is needed;establish more scientific and feasible standards and regulations based on the existing environmental and toxicological data with many differences (due to differences in chemical compounds,species,gender and even strains),by seeking common points while reserving difference.With the update of the understanding of the PFOS issue,there will be feasible ways to solve the harm of PFOS to the environment and human health in the near future.